In an impedance storage device, different impedance statuses of a medium are used to separately indicate information “0” and “1”. However, most impedance storage devices have a write lifetime, and frequent write operations may cause occurrence of a stuck-at fault.
When a hardware error stuck-at fault occurs in an impedance storage device, a storage unit in which the error occurs keeps a fixed value “0” or “1”, and the value of the unit cannot be further rewritten forever. In addition, a probability of stuck-at fault occurrence is much higher than that of a transient error, and the stuck-at fault becomes a major fault error in an impedance memory.
Although a multi-bit error correction mechanism based on Hamming code in an existing memory system can be used to correct this type of stuck-at fault, the multi-bit error correction mechanism based on the Hamming code is write-intensive, which may increase stuck-at fault occurrence in the system.